Traditionally, one or more test prints must be made and evaluated to determine the proper exposure time and color filtration required to properly expose and color balance a print of an image on particular positive or negative film. More recently, with the introduction of color filter calculator devices such as, for example, the Subtractive Calculator Kit by Beseler and the Mitchell Duocube by Unicolor, the determination of exposure time and color filtration has allegedly been simplified. Typically with such devices one must make a test print through the calculator using light transmitted through the frame to be printed and a diffuser. After processing the print, the calculator pattern of a multitude of dots or areas is compared against a standard gray card to determine the proper color filtration. The exposure time may be determined by the dot or area which fixes the color filtration or by an area or dot in another pattern of area produced through the calculator. If the calculator print produces no acceptable time or filtration, a second such print with initial filter and/or time parameters may be required. Nevertheless, even if the first calculator test print is sufficient to give time and filtration, those parameters suffice only for that frame of that film on that paper. If another frame is to be printed, the whole process is repeated. In some cases, however, it is suggested that if a properly exposed frame with good color balance is included in the roll of film with which you are printing, the color filtration and exposure time obtained for that frame may be used as at least a starting point for the other frames. Thus even with these color calculator devices, there is still required a great deal of experimentation from frame to frame regarding color balance and exposure time.
Producers of film and paper usually suggest a color filtration but this can not be relied on as an optimum because of the variations in each individual's enlarger light, lens, age of the lamp, voltage levels, film, and paper age at processing and other variables.